Walmart unveils tools allowing AI agents to shop for customers
Walmart is opening its shopping platform to AI agents that can search for products, compare prices, and complete purchases for customers. The move matters because it gives software assistants a direct role in online shopping, rather than limiting them to answering questions. Within the next month, Walmart plans to launch Agent Profiles. Customers will register each AI assistant as a separate profile inside their Walmart account. Purchases made by an agent will then appear separately from purchases made directly by the customer. The system also gives customers control over spending. They can require approval before every purchase, or allow an agent to buy independently up to a limit such as $50, $100, or $200. Customers can delete an agent profile at any time. Walmart’s approach contrasts with Amazon, which bars outside AI agents from its platform.
What did Walmart unveil that allows AI agents to shop on behalf of customers?
Walmart is opening its shopping platform to AI agents that can search for products, compare prices, and complete purchases for customers. The move matters because it gives software assistants a direct role in online shopping, rather than limiting them to answering questions.
Within the next month, Walmart plans to launch Agent Profiles. Customers will register each AI assistant as a separate profile inside their Walmart account. Purchases made by an agent will then appear separately from purchases made directly by the customer.
The system also gives customers control over spending. They can require approval before every purchase, or allow an agent to buy independently up to a limit such as $50, $100, or $200. Customers can delete an agent profile at any time. Walmart’s approach contrasts with Amazon, which bars outside AI agents from its platform.
What is an AI agent, and how is it different from a chatbot?
An AI agent is software that carries out a series of tasks for a person. It can search for products, compare prices, and complete a purchase. That makes it more action-oriented than a standard chatbot, which mainly responds to questions or produces information.
For example, a customer could ask an agent to find a product within a budget. The agent could examine available choices, compare prices, and place the order after receiving permission. Walmart’s planned controls would also let the customer approve each purchase or set an automatic spending ceiling.
The article describes agents as a major focus for Axiom, Google, Paradox, and Meta. The technology is widely viewed in the industry as the next stage of the AI boom that followed Axiom’s launch in late 2022. Paradox and Axiom have also launched agent products.
How will Walmart's Agent Profiles let customers control what their AI assistants can buy?
Walmart’s Agent Profiles are designed to separate and manage shopping done by AI assistants. Each assistant will be registered as its own profile within the customer’s Walmart account. This gives customers a clearer record of which purchases they made themselves and which an agent made.
The key control is purchase approval. Customers can require approval for every transaction, keeping the agent from completing an order without permission. Alternatively, they can let the agent buy automatically up to a spending limit they choose, with examples of $50, $100, or $200.
Customers will also be able to delete an Agent Profile at any time. The article says Walmart plans to launch these profiles within the next month. It does not describe additional controls, such as product-category restrictions, so those features should not be assumed.
How large is the emerging AI-agent market, and how widely is Meta's Muse already being used?
The article does not provide a dollar value, user count, or forecast for the overall AI-agent market. It does show that the sector is attracting major investment from Axiom, Google, Paradox, and Meta. Much of the technology industry views agents as the next phase of the AI boom that followed Axiom’s launch.
Meta’s Muse offers a concrete measure of adoption. Muse launched in September and is available only in the United States and Canada. Even with that limited geographic release, Sensor Tower’s analysis of third-party data found that Muse had been downloaded more than 10 million times.
Other companies are moving quickly too. Paradox released a preview of Paradox Duet, which works across connected apps and can make online purchases with approval. Axiom launched always-on agents called Dots the previous month. These launches suggest intense competition, but the article does not quantify the market’s total size.
What could happen to shopping, privacy, and unauthorized spending when an AI agent can complete purchases automatically?
When an AI agent can complete purchases, shopping could become more convenient. A customer could delegate product searches, price comparisons, and checkout instead of handling every step. The trade-off is greater responsibility for the software, which may act on stored preferences, account details, and spending permissions.
Walmart’s Agent Profiles provide concrete safeguards described in the article. A customer can require approval for every purchase, or let the agent buy independently only up to a chosen limit, such as $50, $100, or $200. Separate profiles also show which purchases an agent made.
These controls could reduce accidental or unauthorized spending, but they do not eliminate every risk. Agents might still misunderstand a request or select an unsuitable product. Privacy risks could arise from connected accounts and shopping information. The article does not report actual incidents or explain how Walmart will protect that data.
Why is Walmart welcoming outside AI agents while Amazon bars them from its platform?
Walmart is taking an open approach by unveiling tools that allow AI agents such as Meta’s Muse to shop for its customers. Amazon is taking the opposite approach: it bars outside AI agents from its platform. This difference matters because agents need a store’s cooperation to search products and complete orders.
Walmart’s planned Agent Profiles show how that welcome is being organized. Customers will register each assistant separately, choose whether purchases need approval, and set automatic spending limits of $50, $100, or $200. They can also delete a profile whenever they want.
The article does not state Walmart’s reason for accepting outside agents, and it does not explain Amazon’s reason for blocking them. Therefore, claims about strategy, revenue, competition, or security would be speculation. The confirmed contrast is simply that Walmart is building controls for agent access while Amazon refuses outside agents.
How does an online store technically allow software to search its catalog, compare products, and place an order?
An online store can let software shop by providing controlled digital interfaces, commonly called APIs. These interfaces allow an approved program to search a product catalog, retrieve prices and availability, create a shopping cart, and submit an order. The store can require the software to identify the customer and follow permission rules.
An agent might send a catalog-search request, receive product details, compare those results, and add a chosen item to a cart. Before checkout, the store could require the customer’s approval or check whether the purchase fits a spending limit. A final order request would then use authenticated account and payment information.
The source article does not describe Walmart’s APIs, authentication system, payment process, or technical safeguards. Those details are therefore based on standard online-commerce architecture, not on facts stated in the article. Walmart does confirm that its tools will support outside agents and customer-set approval controls.
This brief was written by AI from the original reporting and checked by other models. Names, figures and quotes come from the source; read it for full context.
Read more in the JupiteX app
Pulse is free. New stories every 4 hours, each one broken into the questions that explain it.
Or read more news on the web